Content uploaded by Raul Gomez
Author content
All content in this area was uploaded by Raul Gomez on Feb 27, 2015
Content may be subject to copyright.
[This is the accepted version of a manuscript published in Irish Political Studies.
http://dx.doi.org/10.1080/07907184.2015.1023716]
Shared identity and the end of conflict? How far has a common sense of ‘Northern Irishness’
replaced British or Irish allegiances since the 1998 Good Friday Agreement?
Jonathan Tonge (University of Liverpool)
Raul Gomez (University of Derby)
Abstract
Despite political progress in Northern Ireland, the polity may arguably only fully stabilise when
its population regards themselves as ‘Northern Irish’ rather than merely as sub-sets of British and
Irish parent nations. Power-sharing and relative peace since the 1998 Good Friday Agreement may
have offered the possibility of the development of a common Northern Irish identity, to allow
consolidation of a political entity challenged by sections of the nationalist minority since its
formation in 1921. Alternatively, the consociational nature of the Agreement may have legitimised
‘separate but equal’ identity politics constructed on the British versus Irish faultline. This articles
tests whether there has been a significant growth of cross-community Northern Irishness since the
Agreement, capable of eroding inter-communal rivalry.
Introduction and Background
The successful reintroduction of devolved power-sharing in Northern Ireland following the 2006
St Andrews Agreement has led to a more secure and consensual dispensation than was ever evident
during the previous unhappy eighty-five years of the polity’s existence. The 2006 Agreement
adjusted, but did not radically alter, the 1998 Good Friday, or Belfast, Agreement, backed via a
referendum and which created the institutional framework within which British Unionists and Irish
nationalists would share power in an Assembly and governing Executive.
The Good Friday Agreement was bereft of mention of Northern Irishness. Instead, it recognised
‘the birthright of all the people of Northern Ireland to identify themselves and be accepted as
British or Irish, or both’.
1
As a consociational deal, the Good Friday Agreement legitimised rival
1
HM Government (1998) The Agreement, para 1 vi.
[This is the accepted version of a manuscript published in Irish Political Studies.
http://dx.doi.org/10.1080/07907184.2015.1023716]
Protestant-Unionist-British and Catholic-Irish Nationalist identities rather than attempted their
dissipation. Parity of esteem between two separate but equal communities was the overarching
goal of the Agreement, not the attempted integration of two national communities which remained
politically and culturally divided. This legitimation of identity has led to various criticisms of the
deal as one which merely reinforced Northern Ireland’s ethnic divide.
2
The peace produced by the
Agreement has been imperfect, accompanied by sectarian and ‘dissident’ republican violence.
Concurrently, the politics has seen episodic wrangling between the ethnic bloc political parties,
part of a ‘never-ending peace process’.
3
The Agreement attempted to reconfigure Northern Ireland’s problems, moving them away from
physical conflict and into the realm of cultural contestation. Divisions over symbols or expressions
of identity such as flags and parades have largely displaced political violence around the
constitutional question. These disputes have been evident in controversies over the flying of the
Union flag and the establishment of the Twaddell ‘civil rights camp’ by Loyalists demanding the
Orange Order’s right to walk ‘traditional’ routes. Assertions of Britishness and its distinctive
Protestant ‘Orange’ form in parts of Northern Ireland and expressions of Irishness are hardly new,
but have assumed greater importance since the Good Friday Agreement was reached. Cultural
contests have been evident as Orange Order marches, erstwhile ‘rituals of state’ but now more
marginalised (Bryan 2000: 60) have been challenged by sections of the Nationalist community
amid the bi-nationalism enshrined in the contemporary political framework. Although the Union
2
See e.g. Wilford, R. and Wilson, R. (2006) The Trouble with Northern Ireland, Dublin: New Island.
3
O’Kane, E. (2013) ‘The Perpetual Peace Process? Examining Northern Ireland’s Never-ending, but
Fundamentally Altering Peace Process’, Irish Political Studies, 28.4, 515-35.
[This is the accepted version of a manuscript published in Irish Political Studies.
http://dx.doi.org/10.1080/07907184.2015.1023716]
appears safe under the constitutional framework of the Good Friday Agreement, Unionism appears
more fragmented and insecure.
Arguably, cultural polarity might only be displaced when most of the 1.7 million inhabitants of
the country adopt a common Northern Irish identity. Without a common allegiance to a shared
cross-community identity, zero-sum game cultural contestation may continue to disfigure
Northern Ireland. Thus far, neither community has proved capable of arriving at non-contentious
or neutral expressions of cultural identity, not least because the political leaderships of both built
their success partly upon strident assertions of a particularistic identity. Whilst controversies over
assertions of rival identities may be containable, with most non-violent, they are also routinized
rejections of the concept of a new Northern Ireland in which older affiliations are displaced by a
deeper sense of commonality. Moreover, violent antagonism around assertions of identity is still
evident, most notably at the height of Northern Ireland’s ‘marching season’. Only with the erosion
of identity polarisation is Northern Ireland likely to develop fully as a political entity in its own
right. The current political model of separate but equal communities does not dissipate cultural
polarity.
Some evidence of growth in Catholic contentment with Northern Ireland’s place in the United
Kingdom, alongside traditional Protestant enthusiasm for the Union, may mean there is only
modest support for a united Ireland. The logical corollary of acceptance of Northern Ireland might
be the adoption of Northern Irishness. As the territorial space becomes de-contested, its sharing
could be followed by mutuality of identity. Some have already tentatively indicated this might be
[This is the accepted version of a manuscript published in Irish Political Studies.
http://dx.doi.org/10.1080/07907184.2015.1023716]
occurring
4
, whilst others have been more bullish. Indeed among the latter camp, the regional press
hailed ‘the rise of the Northern Irish’.
5
Drawing upon evidence from the annual Northern Ireland
Life and Times surveys of identity, the only annually-conducted longitudinal assessment of shifts
in Northern Irish identity, this article assesses whether there has indeed been a growth of Northern
Irishness, or whether traditional polarity remains as pervasive as ever.
‘Northern Irishness’ as an identity amid continuing bi-communalism
The other two devolved entities within the UK already had a strong sense of national identity even
prior to the introduction of devolution at the close of the C20th. Scottishness comfortably exceeds
Britishness as the primary national identification and, by a closer margin, Welshness does likewise,
with Scotland and Wales ascribed the label of country
6
The label of Northern Ireland as a ‘country’
is more contentious. The ‘our wee country’ description afforded the entity by Loyalists is not
shared by Nationalists, who tend to see it primarily as part of the island of Ireland. For recognition
as a ‘country’, cross-community Northern Irishness may be an essential pre-requisite, an indication
of shared allegiance to a distinctive entity which is more than merely a region.
Despite a much longer history of devolution, Northern Ireland has been a failure of country-
building, the partisan Protestant Unionist parliament of 1921-72 failing to attract the allegiance of
the alienated Catholic Nationalist minority. Identities within Northern Ireland are not primordial,
4
Hayes, B. and McAllister, I. (2013) Conflict to Peace: Politics and Society in Northern Ireland Over
Half A Century, Manchester: Manchester University Press.
5
Belfast Telegraph, 11 December 2011: 1.
6
Scully, R. (2013) ‘More Scottish than Welsh? Understanding the 2011 Devolved Elections in Scotland
and Wales’, Regional and Federal Studies, 23.5, 591-612.
[This is the accepted version of a manuscript published in Irish Political Studies.
http://dx.doi.org/10.1080/07907184.2015.1023716]
however, and have switched. Rose’s pioneering work showed that, in the late 1960s, Protestant
identity was reasonably evenly divided between ‘Irish’ (20 per cent) ‘Ulster’ (32 per cent) and
‘British’ (39 per cent).
7
The generational effect wrought by the Troubles diminished a sense of
Irish identity held by some Protestants, whilst the lack of institutional recognition of an Ulster
identity, with Northern Ireland’s parliament suspended, also caused the diminution of regional
associations. Britishness became the dominant Protestant identity amid militant assertions of
Irishness from the Catholic minority. The old Ulster identity held a bridging element across the
ethno-religious divide. Eighty-one per cent of Catholics and 67 per cent of Protestants felt that
‘Ulstermen (sic) of the opposite religion [were] about the same’.
8
Developing a shared sense of Northern Irishness remains problematic in the pluralist post-1998
version of devolution, the legacy of conflict shaping a continuing polarity of Britishness versus
Irishness as rival forms of nationalism and culture. Both communities continue to regard
themselves as belonging to nations of which Northern Ireland forms a part. Nationalists have
always preferred an Irish identity, regardless of whether or not they reject the permanent existence
of Northern Ireland as a political entity. Limited acceptance of the ‘Northern Irish’ label is
predominantly of the basis of its Irish dimension rather than through a strong association with the
northern state.
Unionists also see themselves predominantly as a subset of the wider British population, whose
identity and aspirations they share. The basis of Unionism varies between those who see the
relationship with Great Britain as based primarily on rational, civic equality or those who ground
7
Rose, R. (1971) Governing without Consensus: An Irish Perspective, London: Faber and Faber, 208.
8
Ibid. 214.
[This is the accepted version of a manuscript published in Irish Political Studies.
http://dx.doi.org/10.1080/07907184.2015.1023716]
Unionist though in more overt (and to critics, sectarian) expressions of (Protestant) Britishness.
Unionism developed a wing, predominantly located within the Ulster Unionist Party, desirous of
full integration of Northern Ireland within the UK, a project effectively ended by devolution after
1998 and one in any case non-cognisant of the identity of the Irish nationalist minority. In contrast,
another section of the Unionist population has highlighted its regionalism nearly as much as its
Britishness.
9
The political discourse of Loyalism has always emphasised that ‘the Ulster people
are a special people different from the Irish and different from the other people of the UK, but with
a historic sensibility and right to be part of the Union’.
10
This regionalism has been reflected in a
mutual desire of the British government and Northern Ireland’s Unionists and Loyalists to treat
the country as a place apart, the limits of integration apparent in its separate electoral and party
systems.
Yet the taste for full Northern Irish autonomy within Loyalism has always been limited within the
broader Unionist community, which rows back from ultra expressions of Ulster Protestant ethnic
identity in favour of more generalised Britishness
11
(Loughlin 1985: 228-30). Notwithstanding the
possibilities created by devolution, even those sections of the Democratic Unionist Party (DUP)
9
See e.g. Todd, J. (1987) ‘Two Traditions in Unonist Political Culture’, Irish Political Studies, 2.1, 1-26;
Peatling, G. (2004) The Failure of the Northern Ireland Peace Process, Dublin: Irish Academic Press,
129-34.
10
Finlayson, A. (1997) ‘Discourse and Contemporary Loyalist Identity’ in Shirlow, P. and McGovern, M.
(eds) Who Are ‘The People’? Unionism, Protestantism and Loyalism in Northern Ireland, London: Pluto,
72-94: 91.
11
Loughlin, J. (1995) Ulster Unionism and British National Identity since 1885, London: Pinter, 228-30.
[This is the accepted version of a manuscript published in Irish Political Studies.
http://dx.doi.org/10.1080/07907184.2015.1023716]
which harness regionalism tend to emphasise Britishness first. Indeed few members of the DUP
see their identity as Northern Irish, far more (nearly 80 per cent) preferring the British label.
12
.
At elite level, the rejection of the label of Northern Ireland remains steadfast, for now at least,
amongst the political representatives of the region’s main nationalist party, Sinn Fein. This
eschewing of recognition of the country they serve politically exasperates the political leaders of
Unionism, who complain of Sinn Fein’s reluctance to use the term ‘Northern Ireland’ in official
documentation or press statements, instead instructing civil servants to use the North and referring
to the ‘Six Counties’ in internal party discourse.
13
Of the sizeable Northern Irish parties, only
Alliance, as an avowedly non-sectarian party, actively promotes Northern Irishness, as part of its
rejection, as ‘benign apartheid’, of a polity based upon ‘separate but equal’.
14
Yet even Alliance
acknowledges the existence of diversity within the polity and it remains difficult to produce a
definitive sense of inclusive Northern Irishness capable of draining the wider repositories of
loyalty to parent states.
For many within its confines, Northern Ireland is the territory but it is not their ‘homeland’.
15
The
relationship of Unionists and Nationalists to their parent states may be uneasy. It is marked by
12
Tonge, J., Braniff, M., Hennessey, T., McAuley, J. and Whiting, S. (2014) The Democratic Unionist
Party: From Protest to Power, Oxford: Oxford University Press.
13
Ibid.
14
Alliance Party of Northern Ireland (2012) Building a Shared and Prosperous Northern Ireland Through
Tackling the Cost of Division, Belfast: Alliance Party.
15
Barry, J. (2003) ‘National Identities, Historical Narratives and Patron States in Northern Ireland’, in
Linklater, A. and Waller, M. (eds) Political Loyalty and the Nation State, London: Taylor and Francis,
189-205: 193.
[This is the accepted version of a manuscript published in Irish Political Studies.
http://dx.doi.org/10.1080/07907184.2015.1023716]
conditional loyalty on the Unionist side and often accompanied disdain from their British fellow-
citizens, who tend to view Unionists and Nationalists as ‘equally alien’.
16
Similarly, northern
Nationalists, whilst culturally strongly Irish, have felt political estrangement from their ‘parent’
state, the main parties in the Irish Republic, Sinn Fein excepted, declining to contest northern
elections. Both northern communities offer distinctive cultural illustrations of their patron states,
who have acted as somewhat embarrassed and bewildered parents of their client communities. Yet
partial disavowal from those parents has not brought together Northern Ireland’s two communities
on the basis of an alternative mutual Northern Irishness.
Rather than move towards a single community identity, the political framework of Northern
Ireland’s perpetual peace process has been geared towards a more modest project: the successful
adoption of bi-communalism.
17
The identity paradigm of the Good Friday Agreement involves the
acceptance and management of distinct and competitive cultural and political identities, the two
frequently conflated.
Typically, a bi-communal identity model ‘fossilises communities … and hardly augurs well for
community relations’ in emphasising boundaries and reproducing difference.
18
In the Northern
Ireland context, it might reasonably be asked, ‘so what?’ given that the goal of (relative) peace
appears to have been achieved. If that near-peace has been achieved alongside more comfortable
16
Ibid. 194.
17
Nic Craith, M. (2003) Culture and Identity Politics in Northern Ireland, Basingstoke: Palgrave
Macmillan.
18
Nic Craith op cit., 12; Young, I, (1995) ‘Together in Difference: Transforming the Logic of Group
Political Conflict’, in Kymlicka, W. (ed.) The Rights of Minority Cultures, Oxford: Oxford University
Press.
[This is the accepted version of a manuscript published in Irish Political Studies.
http://dx.doi.org/10.1080/07907184.2015.1023716]
expressions of Irishness enjoyed by nationalists, then where is the problem? With few linkages
between the two communities in terms of religion, sport, education or culture, acceptance of bi-
communalism might be seen as prudent, a necessary legitimation of the essentialist identities either
side of a deep faultline. Moreover, Northern Irishness might be seen as a contrived, artificial
identity, a false amalgamation of much more distinctive Irish or British nationhood. In response,
we note Benedict Anderson’s famous assertion that all communities larger than primordial face-
to-face villages … are imagined
19
and recall the element of common Ulster identity which predated
the Troubles. Bereft of a sense of common identity, Northern Ireland’s citizens may be destined to
endure the benign apartheid of mildly antagonistic cultural and political relations, with the
(admittedly slight) attendant risk of re-ignition of hostilities. Singular identification may be
required to displace current affiliations marked by suspicion of the other.
Given the existence of two distinct communities, it is difficult to produce symbols to represent a
common Northern Irishness. The United States diplomat, Richard Haass, who brokered all-party
talks in 2013 in an attempt to resolve disputes over flags, parades and how to deal with the past,
suggested that political parties design a new Northern Irish flag. This would have added to the
modest progress made in designing neutral symbols, already produced, (but notably not by the
political parties) for the Northern Ireland Assembly and the Police Service of Northern Ireland.
However, the lack of consensus (the parties could not agree on a new flag design) highlighted the
difficulty of framing a neutral Northern Irishness.
20
Indeed the Haass talks were criticised more
19
Anderson, B. (2011) Imagined Communities: Reflections on the Origins and Spread of Nationalism,
London: Verso.
20
‘
Dr Richard Haass raises idea of potential new NI flag, at http://www.bbc.co.uk/news/uk-northern-
ireland-25198298, last accessed 12 January 2014. For the full Haass proposals, see ‘Proposed Agreement
31 December 2013, at http://www.northernireland.gov.uk/haass.pdf, last accessed 12 January 2014.
[This is the accepted version of a manuscript published in Irish Political Studies.
http://dx.doi.org/10.1080/07907184.2015.1023716]
broadly for being ‘informed by a liberal emphasis on institutions being a kind of panacea to ethno-
factional politics’, incapable of dissolving rival traditions and historical narratives and risking their
further legitimation.
21
In aspects of culture, there have been some attempts to recognise the sporting pursuits of the
‘other’ community, the First Minister attending a Gaelic Athletic Association match and the Sinn
Fein culture minister watching the Northern Ireland soccer team. As has been pointed out,
however, even ‘acculturation, the sharing of culture, does not necessarily lead to assimilation, a
sharing of identity, in Northern Ireland or elsewhere’.
22
This interpretation suggests the continued
existence of two separate, if now equal, ethnic communities within Northern Ireland – a state
without its own people in effect - in perpetuity.
Demography and identity: the 2011 Census and tNorthern Ireland Life and Times surveys
Our data analysis utilises the Northern Ireland Life and Times Surveys, as an annual study identity
which has historically required respondents to make a single choice. We use the Life and Times
surveys as they are a very regular annual measure of shifts in identity. In contrast, election surveys,
21
Edwards, A. and McGrattan, C. (2014) ‘When the guns fall silent, the battle over the past begins’,
http://www.discoversociety.org/2014/03/04/when-the-guns-fall-silent-the-battle-over-the-past-begins/,
last accessed 22 May 2014: 1.
22
McGarry, J. and O’Leary, B. (1995) Explaining Northern Ireland, Oxford: Blackwell, 357.
[This is the accepted version of a manuscript published in Irish Political Studies.
http://dx.doi.org/10.1080/07907184.2015.1023716]
although utilising questions on identity, have been more piecemeal, with no surveys conducted for
the 2003 and 2007 Assembly and 2005 General elections.
Regarding the demographic composition of the population in Northern Ireland, the 2011 census
shows a growing proportion of Catholics: 40.8 per cent of the population compared to 41.6 per
cent Protestants. Moreover, 75 per cent of new arrivals in Northern Ireland from the 12 new
entrants to the EU during the latest census decade were Catholic, whilst of the total of 202,000
people resident in Northern Ireland who were not born there (11 per cent of the population, up
from 9 per cent in 2001) half are Catholics and only one-third Protestants.
23
Immigration from
outside the UK and the Republic of Ireland has however only had a marginal effect on the growth
of the Catholic community as a whole (immigrants only represent 5 per cent of Catholics and 2
per cent of Protestants). The main explanation for the growth of Catholics can be found elsewhere:
a majority of schoolchildren in Northern Ireland are Catholic
24
; and 44 per cent of the region's
Catholics are aged below 30, compared to only 34 per cent of Protestants according to census data.
These demographic trends present a challenge for Unionist parties. Moreover, they add to the sense
of Protestant insecurity, population and territorial retreat exacerbating the feeling of a community
in decline, whilst struggling to legitimise a localised British cultural identity often eschewed or
viewed with bafflement elsewhere in the UK.
Census data might conceivably represent an alternative source of identity data, but, perhaps
surprisingly, the 2011 Northern Ireland census was the first to ask questions of national identity.
23
Irish Times, 17 May 2013.
24
Belfast Telegraph, 19 October 2011
[This is the accepted version of a manuscript published in Irish Political Studies.
http://dx.doi.org/10.1080/07907184.2015.1023716]
It allowed respondents to choose as many national identities as they wished. A majority favoured
only one identity box. Sixty-eight per cent of Protestants chose a ‘British only’ national identity,
identical to the findings from the Life and Times survey. Also in the census, only 14.5 per cent of
Protestants chose Northern Irish as a sole identity. However, 25.5 per cent of Protestants ticked
either Northern Irish or both British and Northern Irish (in the 2012 Life and Times survey, 24.3
per cent of Protestants chose the Northern Irish identity). Amongst Catholics, 53 per cent identified
as Irish only, with only 2 per cent of Protestants identifying likewise. Only 10.3 per cent of
Catholics identified as 'British only'.
25
Older age groups in Northern Ireland are more likely to identify as British and younger age groups
are more likely to be Irish, but this may partly reflect a growth in Catholic membership of the
younger strata of Northern Ireland’s population. In the 2011 census the proportion of people with
a British-only national identity ranged from 35 per cent of those aged 0-34 to 50 per cent of those
aged 65 and over. In contrast, those with an Irish-only national identity had a younger age
distribution, ranging from 28 per cent of those aged 0-34 to 18 per cent of those aged 65 and over.
There is, however, little variation by age group among those with a Northern Irish only national
identity.
Finally, the 2011 census shows a fairly similar proportion of people from the main communities
with a Northern Irish identity in some of its variations (Northern Irish only or Northern Irish and
some other identity) amounting to 31 per cent of Catholics, compared to 27 per cent of Protestants.
25
Northern Ireland Executive (2011) ‘Key Census Statistics’, at
http://www.northernireland.gov.uk/index/media-centre/news-departments/news-dfp/news-releases-dfp-
december-2012/news-dfp-111212-census-2011-key.htm,
last accessed 11 January 2015.
[This is the accepted version of a manuscript published in Irish Political Studies.
http://dx.doi.org/10.1080/07907184.2015.1023716]
Exceptionally, the Life and Times survey was not conducted in 2011, but the 2010 data also
indicate very marginal differences between Catholics and Protestants at the beginning of the 2010s
(26 per cent of Catholics considered themselves Northern Irish, compared to 29 of Protestants;
very importantly, the difference between both figures is not statistically significant). However, the
2011 census only allows for a static analysis of the Northern Irish case. What now needs to be
examined are the trends within the main communities within that total population.
Northern Irishness since 2000
Figure 1 shows the basic trends in British, Irish and Northern Irish identity for 2000-2012. The
Northern Ireland Life and Times survey is available since 1998 but data from 1998 and 1999 are
not included because the questions differed. Moreover, devolved power-sharing did not commence
until 1999, so 2000 represents the logical starting point in terms of full years since the inception
of Northern Ireland’s Executive and Assembly (although these were of course suspended from
Autumn 2002 until Spring 2007).
As can be appreciated in Figure 1, British identity has declined since 2000 even though 2012
represents an increase on the immediately preceding years. Conversely, both Irish and Northern
Irish identity have increased since 2000. Given that fieldwork for the 2012 survey was conducted
between 1st October 2012 and 10th January 2013, the decrease in Northern Irish identifiers in that
particular survey (and the corresponding increase in Britishness and Irishness) was possibly
influenced by the polarization amidst the furore (riots and protests) over the removal of the Union
flag from permanent position atop Belfast City Hall.
[This is the accepted version of a manuscript published in Irish Political Studies.
http://dx.doi.org/10.1080/07907184.2015.1023716]
Figure 1 Evolution of British, Irish and Northern Irish identities over time
If we look at the evolution of Northern Irishness for the two main communities (Figure 2), it is
evident that most Protestants started declaring themselves Northern Irish after the first half of the
2000s while the opposite trend is observed among Catholics. By 2006 (the year of the St Andrews
Agreement which finally embedded secure power-sharing), there are already more Northern Irish
identifiers among Protestants than among Catholics according to the Northern Ireland Life and
Times Survey. Although a dip in Northern Irishness can be observed in both communities in 2012,
it was much larger among Catholics. Among Protestants, and in spite of the flags disturbances, the
proportion of Northern Irish identifiers has not returned to the levels found before 2007.
20% 30% 50%40%
2000
2002
2004
2006
2008
2010
2012
2001
2003
2005
2007
2009
year
British Irish Northern Irish
[This is the accepted version of a manuscript published in Irish Political Studies.
http://dx.doi.org/10.1080/07907184.2015.1023716]
Figure 2 Evolution of Northern Irishness over time for Catholics and Protestants
But how have these changes come about? There are two possible answers. One is period effects,
meaning that Protestants may have become fonder of a Northern Irish identity since the 2006 St
Andrews Agreement and the stabilising of political arrangements. This identity adoption may have
been adversely affected, but not entirely reversed, by the 2012 flags riots. An interlinked
explanation is the generational effect, as Northern Irishness may have developed mostly among
new generations of Protestants raised amid ceasefires and largely unaffected by the old violence.
Protestants may be comfortable with the new dispensation and feel that their Britishness is less
challenged amid a considerable diminution of violence.
15% 20% 25% 30% 35%
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2012
year
Protestants Catholics
[This is the accepted version of a manuscript published in Irish Political Studies.
http://dx.doi.org/10.1080/07907184.2015.1023716]
When identity figures are disaggregated by religious community and age groups, trends seem to
support both hypotheses. During the last decade, younger Protestants have been consistently more
likely than older counterparts to declare themselves Northern Irish (Table 1) and less likely to
declare themselves British, and age differences by 2012 are even larger when it comes to British
identity (20 points lower amongst younger Protestants than amongst the oldest age group).
Unfortunately, over only 12 years, it is not possible to assess whether the effect of age is due to
life course effects or whether it reflects generational differences likely to remain over time. It is
possible that younger generations which have been mainly socialised after the Troubles have
developed different identities.
Table 1 Identity amongst Protestants in Northern Ireland 2000-2012, by age
(Sampling weights applied)
Identity and Year 18-30 31-45 46-60 61-75 76+ Total
British
2000
64.72 75.98 69.70 76.01 78.85 72.49
British 2012 60.90 58.59 71.74 75.11 80.82 68.36
Irish 2000 3.06 1.78 3.64 4.34 4.49 3.18
Irish 2012 0.75 1.76 3.91 2.22 6.85 2.77
Northern Irish 2000 21.94 14.77 15.35 10.12 5.77 14.70
Northern Irish 2012 35.34 32.16 20.0 20.0 8.22 24.44
Ulster 2000 7.22 5.69 6.87 6.94 5.77 6.51
Ulster 2012 0.0 3.08 0 1.78 4.11 1.58
[This is the accepted version of a manuscript published in Irish Political Studies.
http://dx.doi.org/10.1080/07907184.2015.1023716]
Other 2000 3.06 1.78 4.44 2.6 5.13 3.13
Other 2012 3.01 4.41 4.35 0.89 0 2.93
This contrasts with what has happened with Catholics (Table 2). Age does not seem to have a
strong effect on Northern Irish identity among Catholics, but there may be a period effect as the
percentage of those declaring themselves Northern Irish has been continuously and significantly
decreasing year upon year (from 28 per cent in 2000 down to 16 per cent in 2012). In contrast,
according to the Life and Times surveys, Irish identity among Catholics has increased from nearly
60 per cent in 2000 to nearly 68 per cent in 2012.
Table 2 Identity amongst Catholics in Northern Ireland 2000-2012, by age
(Sampling weights applied)
Identity and Year
18-30 31-45 46-60 61-75 76 Total
British 2000 4.02 12.54 9.96 11.05 10 9.51
British 2012 3.98 9.16 10.33 8.39 18.03 8.67
Irish 2000 69.48 53.05 53.02 66.28 68.33 59.83
Irish 2012 72.57 63.36 68.18 68.53 65.57 67.77
Northern Irish 2000 25.7 33.76 31.67 19.77 16.67 28.15
[This is the accepted version of a manuscript published in Irish Political Studies.
http://dx.doi.org/10.1080/07907184.2015.1023716]
Northern Irish 2012 12.39 15.65 17.77 21.68 16.39 16.38
Ulster 2000 0 0 1.42 1.16 5 0.84
Ulster 2012 1.77 0 0 0 0 0.43
Other 2000 0.8 0.64 3.91 1.74 0 1.68
Other 2012 9.29 11.83 3.72 1.4 0 6.75
What matters in determining Northern Irishness?
In order to further investigate the above trends we have run several logistic regressions. The
dependent variable is Northern Irish identity (1) vs. any other identity (0). The first two models in
Table 3 are separate models for Catholics and Protestants. Respondents that are neither Protestant
nor Catholic have been excluded (we will come back to them later in a further analysis)
.
Apart
from demographic and political controls, the models contain two important pieces of information:
age and year dummies to examine period effects.
26
Table 3 Logistic Regression Model for Northern Irish identity amongst Protestants and
Catholics 2000-12
(1) (2)
VARIABLES Protestants Catholics
Age -0.015*** -0.002
(0.002) (0.002)
Gender (=male) 0.074 -0.260***
(0.063) (0.071)
Urbanization 0.035 0.144***
(0.028) (0.029)
26
We opted for using year dummies instead of a continuous variable so the significance of changes from
2000 can be appreciated year by year.
[This is the accepted version of a manuscript published in Irish Political Studies.
http://dx.doi.org/10.1080/07907184.2015.1023716]
Social class (ref: skilled/unskilled worker)
Professional or manager 0.062 0.107
(0.079) (0.090)
Never worked/long-term unemployed -0.208 -0.185
(0.218) (0.212)
Education(ref: no formal qualifications)
GCSE or Equivalent 0.196** 0.344***
(0.087) (0.096)
Higher Education/GCE A 0.480*** 0.258**
(0.091) (0.103)
Degree or higher 0.766*** 0.247**
(0.104) (0.117)
Party ID (ref: no Party ID)
DUP -0.593*** -0.324
(0.087) (0.498)
Sinn Fein -0.869 -0.962***
(0.551) (0.112)
UUP -0.199** 0.260
(0.084) (0.349)
SDLP 0.145 0.013
(0.230) (0.084)
Alliance 0.520*** 0.562***
(0.126) (0.173)
Other 0.218 0.223
(0.167) (0.192)
Period (ref: year 2000)
Year 2001 0.162 -0.008
(0.148) (0.150)
Year 2002 -0.054 -0.108
(0.152) (0.150)
Year 2003 0.549*** -0.086
(0.137) (0.157)
Year 2004 0.207 -0.134
(0.144) (0.157)
Year 2005 0.511*** 0.070
(0.157) (0.173)
Year 2006 0.755*** -0.181
(0.152) (0.172)
Year 2007 0.822*** -0.202
(0.152) (0.171)
[This is the accepted version of a manuscript published in Irish Political Studies.
http://dx.doi.org/10.1080/07907184.2015.1023716]
Year 2008 1.033*** -0.071
(0.149) (0.171)
Year 2009 0.899*** -0.201
(0.152) (0.167)
Year 2010 0.889*** -0.040
(0.153) (0.171)
Year 2012 0.597*** -0.626***
(0.167) (0.184)
Constant -1.341*** -1.289***
(0.175) (0.190)
Observations 8,293 6,010
Robust standard errors in parentheses,
sampling weights applied
*** p<0.01, ** p<0.05, * p<0.1
Let us first focus on the trends in Northern Irishness in relation to time and age effects, indicated
by the year dummies at the bottom of the table. In order to ease interpretation, changes in the
probabilities of adopting a Northern Irish identification from the year 2000 (year of reference) are
shown in Figure 3 for both Catholics and Protestants. Levels of Northern Irishness are only
significantly different from those in 2000 if confidence intervals (grey areas) do not overlap zero.
Results show a significant increase in Northern Irishness over time among Protestants, especially
since 2005 and with very high increases between 2006 and 2010. Even if 2012 represented a small
dip compared to previous years, the probability that a random Protestant adopted a Northern Irish
identity in 2012 was still 9 per cent higher in 2012 than it was in 2000. Indeed only the years 2008
(p<0.01), 2009 and 2010 (p<0.1) present significantly higher levels of Northern Irishness than
those found in 2012.
27
27
We also ran a model (not shown) including random effects by year instead of dummies, and period
effects were significant for Protestants.
[This is the accepted version of a manuscript published in Irish Political Studies.
http://dx.doi.org/10.1080/07907184.2015.1023716]
Figure 3 Changes in the probability of having a Northern Irish identification from year 2000
(for the main two communities)
28
By contrast, period effects are not clear when it comes to Catholics, as the levels of Northern
Irishness in every single year are hardly different from those in 2000 (confidence intervals overlap
zero). The only significant period effect took place in 2012, where there is a large dip in Northern
28
Probabilities have been calculated holding all other variables at their means
.
-.2 -.1 0 .1 .2
Chages in the probability of having a Northern Irishness identity from year 2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2012
year
Point estimates and 95% confidence bounds (grey areas)
Protestants
-.2 -.1 0 .1 .2
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2012
year
Departures from the red line represent changes from levels of Northern Irishness in 2000
Catholics
[This is the accepted version of a manuscript published in Irish Political Studies.
http://dx.doi.org/10.1080/07907184.2015.1023716]
Irishness among Catholics (10 per cent less than in 2000). Whilst for Protestants Northern Irishness
is significantly higher in 2012 than it was in 2000, it is nevertheless lower for Catholics.
We suggested earlier that there are two possible explanations for the increase in Northern Irishness
among Protestants over the past few years. One plausible reason was period effects, and our results
seem to confirm this is part of the explanation. The second reason is generational effects, if younger
Protestants are more likely than older counterparts to embrace a Northern Irish identity. Models
in Table 3 also test for the latter explanation by introducing age as independent variable. In order
to facilitate the interpretation of the coefficients, Figure 4 below shows the marginal effect of age
for both communities. As can be appreciated via the left-hand graph, age clearly has a significant
impact for Protestants, with younger Protestants being more likely than older counterparts to
consider themselves Northern Irish. There is therefore both a period effect, due to the dramatically
changed political context, and an age effect. Age differences have been developing for the last 12
years and continue to do so at the same pace.
29
In contrast, age does not play a significant role
when it comes to Catholics as Northern Irishness is equally present across age groups (right-hand
graph in Figure 4).
29
Interactions between age and periods are not significant.
[This is the accepted version of a manuscript published in Irish Political Studies.
http://dx.doi.org/10.1080/07907184.2015.1023716]
Figure 4 The effect of age on the probability of having a Northern Irish identification
30
To sum up, there has been a gradual reinforcement of Northern Irishness over the past years among
Protestants and most especially among the youth, who present the highest levels of Northern
Irishness in the two communities. Amongst Catholics, the only significant decrease in Northern
Irishness is in 2012. Prior to this, the levels of Northern Irishness had subsided slightly over time
but not at a marked pace and not significantly across particular ages. The flags issue may have had
a backlash effect in this community, rapidly accelerating a gentler trend away from Catholic
30
Probabilities have been calculated holding all other variables at their means.
.15 .2 .25 .3 .35
Probability of having a Northern Irish identification
20 40 60 80
age
Protestants
Point estimates and 95% confidence bounds (grey areas)
20 40 60 80
age
Catholics
[This is the accepted version of a manuscript published in Irish Political Studies.
http://dx.doi.org/10.1080/07907184.2015.1023716]
adoption of a Northern Irish identity since the Good Friday Agreement, amid overt recognition of
their Irishness.
Once the effects of age and time have been disentangled, it is worth looking at some of the other
relevant variables in Table 3 above. In the first place, the effect of gender is only significant for
Catholics (Model 2) but not for Protestants (Model 1). Catholic males are 4.6 per cent less likely
than Catholic women to adopt a Northern Irish identity.
31
This may perhaps be explained by
Nolan's finding that the big winners from the peace process are Catholic middle classes and
Catholic females.
32
Similarly, urbanization only has a significant effect for Catholics. In particular,
Catholics living in big cities are 10 per cent more likely to adopt a Northern Irish identity than
those living in the country. Regarding education, the most educated sections of the Northern Irish
population are those most likely to embrace Northern Irishness. This appears logical, as they may
have benefitted most from the (often good) educational provision and attendant benefits within the
northern state. This effect is significant for both communities, although stronger among
Protestants, amongst whom a higher the level of education equates to a much greater chance of
feeling Northern Irish. Protestants holding a degree or higher qualification are 13 per cent more
likely than those with no qualifications to adopt a Northern Irish identity. When both education
31
All probabilities mentioned in the text have been calculated holding all other variables at their means.
32
Nolan (2012).
[This is the accepted version of a manuscript published in Irish Political Studies.
http://dx.doi.org/10.1080/07907184.2015.1023716]
and urbanization are controlled for, social class
33
is not significant for either of the two
communities.
34
The above models also control for the party identification of respondents, and effects are what one
might expect. Supporters of the ‘big two’, the DUP and Sinn Fein, are hostile to Northern Irishness,
preferring to adopt British and Irish identifications respectively. Although the DUP has stressed a
regional identity, its backers tend to eschew an identity which embraces ‘Irishness’, even in its
milder Northern Irish form. The identity of Sinn Fein’s identifiers reflects their party’s identity,
ethos and politics. As an all-Ireland party which does not recognise the border in terms of its own
organisation and remains committed to the dissolution of the border (albeit via peaceful means)
Sinn Fein promotes Irishness, in terms of language, culture and identification. Backers of the
centrist Alliance like to identify as Northern Irish, reflecting that party’s commitment to internal
integration of the two communities within Northern Ireland.
So far, we have looked at Catholics and Protestants separately. However, we have not done any
test regarding whether the differences reported earlier between the two communities are
statistically significant. If they were not, that would mean that, after all, members of both
communities are just as likely to adopt a Northern Irish identity in spite of the apparent trends
mentioned above.
33
We only include two categories of social class: professionals and managers, and skilled/unskilled
workers (and a separate category for those that have not been employed in the past few years). The way in
which social class is measured in the Northern Ireland Life and Times surveys has changed over time and
so these categories correspond to the lowest common denominator across surveys.
34
Without controlling for those variables, professionals and managers from both communities are
significantly more likely to feel Northern Irish.
[This is the accepted version of a manuscript published in Irish Political Studies.
http://dx.doi.org/10.1080/07907184.2015.1023716]
To investigate this, we have run an additional model including interactions with religious
community (we present the full model in Table 1A in the appendix).
35
Based on this model, Figure
5 shows by how much the chances of declaring oneself Northern Irish change when the respondent
is Protestant rather than Catholic.
36
Thus, positive numbers mean the probability of having a
Northern Irish identity in a given year is higher for Protestants, whereas negative numbers mean it
is higher for Catholics.
37
As can be seen, until 2005 the probability was higher for Catholics. In
2000 a Catholic was 13 per cent more likely than a Protestant to have a Northern Irish identity.
However, during the 2005-2010 period differences between both groups vanish and are not
statistically different until 2012. That year, due to the deep drop in Northern Irishness among
Catholics that we saw before, the chances of finding a Northern Irish identifier are significantly
higher among Protestants (i.e. the confidence intervals do not overlap with the zero line). To be
more precise, in 2012 a Protestant was 7 per cent more likely than a Catholic to have a Northern
Irish identity.
Figure 5 The probability of having a Northern Irish identity: difference between
Protestants and Catholics 2000-12
35
In particular, we interacted religion with age and period dummies, but also education, gender and urbanization as
those variables had previously been reported to have different effects for each community.
36
Probabilities have been calculated holding all other variables at their means.
37
Here we follow Kam and Franzese (2007), who recommend interpreting interactions by testing for differences in
marginal effects.
[This is the accepted version of a manuscript published in Irish Political Studies.
http://dx.doi.org/10.1080/07907184.2015.1023716]
Thus far we have referred to differences between the average Catholic and Protestant. However,
we saw earlier that young Protestants have been consistently more likely than older counterparts
to consider themselves Northern Irish, an effect that is not present for Catholics. This means that
the largest differences between Catholics and Protestants should be found among young people.
This can be clearly seen in Figure 6 below, which shows the difference in the probability of
declaring oneself Northern Irish for Protestants and Catholics at two different ages: 18 years old
(left hand graph) and 55 years old (right hand graph). As can be seen, young Catholics and
Protestants used to exhibit similar levels of Northern Irishness at the beginning of the 2000s,
immediately after the Good Friday Agreement (confidence intervals overlap zero). However,
differences have been continuously growing in the past few years. In 2006, young Protestants were
-.2 -.1 0 .1 .2
Difference in Northern Irishness between Catholics and Protestants over time
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2012
year
Conditional Marginal Effects with 95% confidence bounds (grey area)
[This is the accepted version of a manuscript published in Irish Political Studies.
http://dx.doi.org/10.1080/07907184.2015.1023716]
already 11 per cent more likely than young Catholics to consider themselves Northern Irish
(differences are statistically significant). By 2012, they were 15 per cent more likely.
Differences among middle-aged people, however, follow a pattern similar to the one observed in
Figure 6 for the average Catholic and Protestant. In 2000, 55-years-old Protestants were 14 per
cent less likely than Catholic counterparts to consider themselves Northern Irish, but these
differences started to vanish over the years and by 2006 were no longer significant.
Figure 6 The probability of having a Northern Irish identity: difference between
Protestants and Catholics at two ages (18 and 55 years old).
18 years old
-.2 -.1 0 .1 .2
Northern Irishness (differences between Catholics and Protestants)
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2012
year
Conditional Marginal Effects with 95% confidence bounds (grey areas)
55 years old
-.2 -.1 0 .1 .2
Northern Irishness (differences between Catholics and Protestants)
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2012
year
[This is the accepted version of a manuscript published in Irish Political Studies.
http://dx.doi.org/10.1080/07907184.2015.1023716]
Finally, it is worth paying attention to the non-religious sector of the population. Non-religious
identifiers are growing in Northern Ireland, amounting to 10.1 per cent of the population according
to the 2011 census, which, combined with the 6.8 per cent religion ‘not stated’, produces a 16.9
per cent non-religious group. The 2012 Life and Times Survey found that 15 per cent of
respondents declared they were of no religion. To what extent does this group adopt or eschew a
Northern Irish identity? Table 4 looks at this, and it also shows the results of the model for the
whole sample (that is, including Catholics, Protestants and not religious).
When it comes to non-religious identifiers in Table 4, we can also see a negative effect of age with
Northern Irishness being stronger among the youth. This negative effect of increasing age is
therefore common to Protestants and non-religious identifiers; there is no significant difference
between the size of the effect for both groups. Regarding the effect of time, there is however no
clear pattern affecting non-religious identifiers. Holding age constant, Northern Irish identity is
only as prevalent among non-religious identifiers as it was in 2000; it has barely increased over
the years even as Northern Ireland stabilized.
Looking at the whole population (second model in Table 4), age has a negative effect on the
likelihood of adoption of Northern Irish identification, with older people being significantly less
likely to be labelled as such. But we now know that this is mainly due to generational differences
among both Protestants and non-religious identifiers. There is also a significant effect of time in
the whole population. Northern Irishness increases with time and particularly from 2005 onwards,
an effect mainly led by the Protestant community. However, this increase stopped in 2012 and
Protestant versus Catholic differences became more marked.
[This is the accepted version of a manuscript published in Irish Political Studies.
http://dx.doi.org/10.1080/07907184.2015.1023716]
Table 4
Logistic Regression Model for Northern Irish identity: all groups (combined) and
non-religious 2000-12
VARIABLES Non-religious
identifiers
All
Age -0.012*** -0.009***
(0.004) (0.001)
Gender (=male) -0.040 -0.059
(0.117) (0.043)
Urbanization 0.070 0.093***
(0.054) (0.018)
Social class (ref: skilled/unskilled worker)
Professional or manager -0.188 0.062
(0.147) (0.054)
Never worked/long-term unemployed -0.349 -0.197
(0.430) (0.140)
Education (ref: no formal qualifications)
GCSE or Equivalent 0.407** 0.260***
(0.172) (0.059)
Higher Education/GCE A 0.682*** 0.409***
(0.171) (0.062)
Degree or higher 0.407** 0.478***
(0.192) (0.071)
Party ID (ref: no Party ID)
DUP -0.516** -0.581***
(0.208) (0.070)
Sinn Fein -1.206*** -1.010***
(0.352) (0.096)
UUP -0.221 -0.291***
(0.210) (0.064)
SDLP -0.141 0.092
(0.208) (0.062)
Alliance 0.602*** 0.557***
(0.189) (0.086)
Other 0.296 0.219**
(0.215) (0.105)
Period (ref: year 2000)
Year 2001 0.714*** 0.191**
(0.266) (0.095)
[This is the accepted version of a manuscript published in Irish Political Studies.
http://dx.doi.org/10.1080/07907184.2015.1023716]
Year 2002 -0.373 -0.051
(0.285) (0.098)
Year 2003 0.156 0.264***
(0.298) (0.096)
Year 2004 -0.116 0.032
(0.275) (0.097)
Year 2005 0.365 0.335***
(0.320) (0.106)
Year 2006 0.467* 0.346***
(0.269) (0.103)
Year 2007 0.397 0.354***
(0.294) (0.105)
Year 2008 0.142 0.487***
(0.275) (0.102)
Year 2009 -0.018 0.358***
(0.273) (0.102)
Year 2010 0.339 0.463***
(0.283) (0.104)
Year 2012 0.119 0.065
(0.286) (0.111)
Constant -1.006*** -1.315***
(0.320) (0.116)
Observations 1,828 16,388
Robust standard errors in parentheses, sample weights applied
*** p<0.01, ** p<0.05, * p<0.1
Conclusion
Northern Ireland may have become more secure as a political entity, but as an identity Northern
Irish has failed to flourish on a cross-community basis. Both communities do not look across to
each other to redefine nationality, which is instead shaped by intra-communal debates over how
best to express Britishness or Irishness. The largest political parties do not encourage a common
Northern Irish identity and their supporters take their cue accordingly, the DUP and Sinn Fein
[This is the accepted version of a manuscript published in Irish Political Studies.
http://dx.doi.org/10.1080/07907184.2015.1023716]
support bases both repudiating Northern Irishness in favour of British and Irish identifications
respectively.
Our analysis of the time series data suggests, however, that there are clear signs of younger non-
religious individuals and young Protestants, particularly the well-educated, being significantly
more likely to adopt a Northern Irish identity, even if episodic rows, such as the furore over the
flying of national flags, (an issue unresolved by the Haass talks process) can temporarily harden
British or Irish identities at the expense of Northern Irishness. Generational effects are the sum of
period and age effects that crystallize with time. The fact that young Protestants are now much
more likely to consider themselves Northern Irish is important. There may be an ongoing
generational process that continues to boost Northern Irishness among Protestants in the future.
Older generations lived through the violence of the Troubles as Northern Ireland’s constitutional
status was challenged by an armed group determined to end the Britishness of the region. This
grouping is now being replaced with younger Northern Irish identifiers no longer confronted by
challenges to Northern Ireland’s UK status and thus comfortable with a regional Northern Irish
identity.. Protestants overall are now more likely to identify as Northern Irish compared to
Catholics. Indeed Catholics, who at the time of the Good Friday Agreement were more likely than
Protestants to identify as Northern Irish (the Irish part of that label was a crucial aspect for
Catholics) are now much more reluctant – and the flags issue may have weakened the Northern
Irish identity of Catholics far more than it weakened the Northern Irish identity of Protestants.
A notable achievement of the Good Friday Agreement lies in the Catholic Irish population
accepting that in Northern Ireland they can be comfortably Irish under British rule over the
country. This is the deal’s triumph, rather than the bi-communal adoption of Northern Irishness as
[This is the accepted version of a manuscript published in Irish Political Studies.
http://dx.doi.org/10.1080/07907184.2015.1023716]
an identity which ‘non-sectarian’ parties such as Alliance desire as necessary for the acute
divisions of the past to be truly overcome. It has been claimed that we have ‘witnessed a
transformation in identities’ in a more peaceful Northern Ireland.
38
Yet change has been confined
mainly to the downplaying of the more contentious history and myth associated with those
identities. It does not extend substantially to a new national identification which moves beyond
the old British versus Irish groupings, with Catholics increasingly eschewing Northern Irishness
whereas Protestants have become more accepting of the label. The old zero-sum game of
Britishness versus Irishness may have been partly replicated in terms of Northern Irishness. As
younger Protestants have embraced, Catholics have become less keen. The Catholic Irish have
come of age within the northern state. Having had expressions of that Irishness banned for several
generations (even the flying of the tricolour was banned for a period) the Irishness of the minority
community has been legally and politically endorsed since the Good Friday Agreement. Irishness
is thus seen as the natural identity to adopt, but such identification does not automatically equate
to demands for a change in Northern Ireland’s constitutional status, many of that Catholic Irish
population enjoying participation within the northern state for the first time.
39
The extent of internal transformation, in terms of diversity and plurality within those identities can
be overplayed. Britishness and Unionism remain at the heart of the majority community, even if
Protestantism has diminished somewhat.
40
Irishness, Nationalism and Catholicism are still
38
Walker, B. (2012) A Political History of the Two Irelands, Basingstoke: Palgrave Macmillan, 203.
39
Evans, J. and Tonge, J. (2013) ‘From Abstentionism to Enthusiasm: Sinn Fein, Nationalist Electors and
Support for Devolved Power-Sharing in Northern Ireland’, Irish Political Studies, 2013, 28.1, 39-57.
40
Farrington. C. (2006) Ulster Unionism and the Peace Process in Northern Ireland, Basingstoke;
Palgrave Macmillan.
[This is the accepted version of a manuscript published in Irish Political Studies.
http://dx.doi.org/10.1080/07907184.2015.1023716]
important features of the communal identity of the minority community.
41
This is not to deny the
emergence of alternative identifications at the sub-national level as territorial and constitutional
questions subside
42
(see Hancock 2013) but these may not be framed within the context of a new
and shared Northern Irishness – the common identity which might finally secure Northern Ireland
as a healthy functioning polity rather than a site of ethnic antagonisms. Until that point is reached,
it will remain a divided, bi-communal entity with a deep political and cultural faultline built upon
rival national affiliations: Irish, with some, but diminishing Northern Irishness, on the Catholic
side; British, with a modest rise in Northern Irishness amongst Protestants. Under this rival
identification scenario, each community’s ethno-national and ethno-religious symbols and markers
may continue to outweigh a broader sense of a national community.
41
Evans, J. and Tonge, J. (2013) ‘Catholic, Irish and Nationalist: evaluating the importance of ethno-
national and ethno-religious variables in determining nationalist political allegiance in Northern Ireland’,
Nations and Nationalism, 19.2, 357-75.
42
Hancock, L. (2013) ‘Peace from the People: Identity Salience and the Northern Irish Peace Process’, in
White, T. (ed.) (2013) Lessons from the Northern Ireland Peace Process, Madison, Wisconsin: University
of Wisconsin Press, 61-93.
[This is the accepted version of a manuscript published in Irish Political Studies.
http://dx.doi.org/10.1080/07907184.2015.1023716]
References
Alliance Party of Northern Ireland (2012) Building a Shared and Prosperous Northern Ireland
Through Tackling the Cost of Division, Belfast: Alliance Party.
Anderson, B. (2006) Imagined Communities: Reflections on the Origin and Spread of Nationalism,
London: Verso.
Barry, J. (2003) ‘National Identities, Historical Narratives and Patron States in Northern Ireland’,
in Linklater, A. and Waller, M. (eds) Political Loyalty and the Nation State, London: Taylor and
Francis, 189-205:
Bryan, D. (2000) Orange Parades, London: Pluto.
Edwards, A. and McGrattan, C. (2014) ‘When the guns fall silent, the battle over the past
begins’, http://www.discoversociety.org/2014/03/04/when-the-guns-fall-silent-the-battle-over-
the-past-begins/, last accessed 22 May 2014: 1.
Evans, J. and Tonge, J. (2013a) ‘From Abstentionism to Enthusiasm: Sinn Fein, Nationalist
Electors and Support for Devolved Power-Sharing in Northern Ireland’, Irish Political Studies,
2013, 28.1, 39-57.
Evans, J. and Tonge, J. (2013b) ‘Catholic, Irish and Nationalist: evaluating the importance of
ethno-national and ethno-religious variables in determining nationalist political allegiance in
Northern Ireland’, Nations and Nationalism, 19.2, 357-75.
Farrington. C. (2006) Ulster Unionism and the Peace Process in Northern Ireland, Basingstoke;
Palgrave Macmillan.
Finlayson, A. (1997) ‘Discourse and Contemporary Loyalist Identity’ in Shirlow, P. and
McGovern, M. (eds) Who Are ‘The People’? Unionism, Protestantism and Loyalism in Northern
Ireland, London: Pluto, 72-94.
Hancock, L. (2013) ‘Peace from the People: Identity Salience and the Northern Irish Peace
Process’, in White, T. (ed.) (2013) Lessons from the Northern Ireland Peace Process, Madison,
Wisconsin: University of Wisconsin Press, 61-93.
HM Government (1998) The Agreement, Belfast: HMSO.
Kam, C. and Franzese, R. (2007) Modeling and Interpreting Interactive Hypotheses in
Regression Analysis, Ann Arbor; University of Michigan Press.
[This is the accepted version of a manuscript published in Irish Political Studies.
http://dx.doi.org/10.1080/07907184.2015.1023716]
Loughlin, J. (1995) Ulster Unionism and British National Identity since 1885, London: Pinter.
McGarry, J. and O’Leary, B. (1995) Explaining Northern Ireland, Oxford: Blackwell.
Nic Craith, M. (2003) Culture and Identity in Northern Ireland, Basingstoke: Palgrave Macmillan.
Nolan, Paul. 2012. Northern Ireland Peace Monitoring Report, 1. Belfast: Community Relations
Council.
O’Kane, E. (2013) ‘The Perpetual Peace Process? Examining Northern Ireland’s Never-ending,
but Fundamentally Altering Peace Process’, Irish Political Studies, 28.4, 515-35.
Peatling, G. (2004) The Failure of the Northern Ireland Peace Process, Dublin: Irish Academic
Press.
Porter, N. (1996) Rethinking Unionism, Belfast: Blackstaff.
Rose, R. (1971) Governing without Consenus: An Irish Perspective, London: Faber & Faber.
Scully, R. (2013) ‘More Scottish than Welsh? Understanding the 2011 Devolved Elections in
Scotland and Wales’, Regional and Federal Studies, 23.5, 591-612.
Todd, J. (1987) ‘Two Traditions in Unonist Political Culture’, Irish Political Studies, 2.1, 1-26.
Tonge, J., Braniff, M., Hennessey, T., McAuley, J. and Whiting, S. (2014) The Democratic
Unionist Party: From Protest to Power, Oxford: Oxford University Press.
Walker, B. (2012) A Political History of the Two Irelands, Basingstoke: Palgrave Macmillan.
Wilford, R. and Wilson, R. (2006) The Trouble with Northern Ireland, Dublin: New Island.
Young, I, (1995) ‘Together in Difference: Transforming the Logic of Group Political Conflict’,
in Kymlicka, W. (ed.) The Rights of Minority Cultures, Oxford: Oxford University Press.
[This is the accepted version of a manuscript published in Irish Political Studies.
http://dx.doi.org/10.1080/07907184.2015.1023716]
APPENDIX
Table 1A
Logistic Regression Model for Northern Irish identity (main communities) 2000-12
(3)
VARIABLES Full Model
with
Interactions
Protestant -0.048
(0.255)
Age -0.002
(0.002)
Gender (=male) -0.258***
(0.071)
Urbanization 0.145***
(0.029)
Education(ref: no quals)
GCSE or Equivalent 0.351***
(0.095)
Higher Education/GCE A 0.267***
(0.101)
Degree or higher 0.266**
(0.112)
Period (ref: year 2000)
Year 2001 -0.012
(0.150)
Year 2002 -0.110
(0.150)
Year 2003 -0.092
(0.156)
Year 2004 -0.138
(0.156)
Year 2005 0.065
(0.173)
Year 2006 -0.179
(0.172)
Year 2007 -0.199
(0.170)
[This is the accepted version of a manuscript published in Irish Political Studies.
http://dx.doi.org/10.1080/07907184.2015.1023716]
Year 2008 -0.076
(0.171)
Year 2009 -0.207
(0.166)
Year 2010 -0.047
(0.170)
Year 2012 -0.629***
(0.184)
Protestant x Age -0.013***
(0.003)
Protestant x Male 0.329***
(0.094)
Protestant x Urbanization -0.111***
(0.040)
Protestant x GCSE -0.159
(0.128)
Protestant x Higher Education 0.208
(0.134)
Protestant x Degree or higher 0.489***
(0.143)
Protestant * Year 2001 0.175
(0.211)
Protestant * Year 2002 0.056
(0.214)
Protestant * Year 2003 0.641***
(0.207)
Protestant * Year 2004 0.344
(0.212)
Protestant * Year 2005 0.444*
(0.233)
Protestant * Year 2006 0.931***
(0.228)
Protestant * Year 2007 1.019***
(0.227)
Protestant * Year 2008 1.107***
(0.226)
Protestant * Year 2009 1.112***
(0.223)
Protestant * Year 2010 0.942***
(0.226)
Protestant * Year 2012 1.227***
(0.248)
Social class (ref: skilled/unskilled worker)
Professional or manager 0.083
[This is the accepted version of a manuscript published in Irish Political Studies.
http://dx.doi.org/10.1080/07907184.2015.1023716]
(0.059)
Never worked/long-term unemployed -0.193
(0.154)
Party ID (ref: no PID)
DUP -0.579***
(0.083)
Sinn Fein -0.968***
(0.107)
UUP -0.178**
(0.078)
SDLP 0.009
(0.076)
Alliance 0.536***
(0.102)
Other 0.222*
(0.126)
Constant -1.292***
(0.189)
Observations 14,303
Robust standard errors in parentheses, samples weights applied
*** p<0.01, ** p<0.05, * p<0.1
